The caches used to store sunrpc authentication information can be
flushed by writing a timestamp to a file in /proc.
This timestamp has a one-second resolution and any entry in cache that
was last_refreshed *before* that time is treated as expired.
This is problematic as it is not possible to reliably flush the cache
without interrupting NFS service.
If the current time is written to the "flush" file, any entry that was
added since the current second started will still be treated as valid.
If one second beyond than the current time is written to the file
then no entries can be valid until the second ticks over. This will
mean that no NFS request will be handled for up to 1 second.
To resolve this issue we make two changes:
1/ treat an entry as expired if the timestamp when it was last_refreshed
is before *or the same as* the expiry time. This means that current
code which writes out the current time will now flush the cache
reliably.
2/ when a new entry in added to the cache - set the last_refresh timestamp
to 1 second *beyond* the current flush time, when that not in the
past.
This ensures that newly added entries will always be valid.
Now that we have a very reliable way to flush the cache, and also
since we are using "since-boot" timestamps which are monotonic,
change cache_purge() to set the smallest future flush_time which
will work, and leave it there: don't revert to '1'.
Also disable the setting of the 'flush_time' far into the future.
That has never been useful and is now awkward as it would cause
last_refresh times to be strange.
Finally: if a request is made to set the 'flush_time' to the current
second, assume the intent is to flush the cache and advance it, if
necessary, to 1 second beyond the current 'flush_time' so that all
active entries will be deemed to be expired.
As part of this we need to add a 'cache_detail' arg to cache_init()
and cache_fresh_locked() so they can find the current ->flush_time.
Signed-off-by: NeilBrown <neilb@suse.com>
Reported-by: Olaf Kirch <okir@suse.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Switch using list_head for cache_head in cache_detail,
it is useful of remove an cache_head entry directly from cache_detail.
v8, using hash list, not head list
Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Nfsd has implement a site of seq_operations functions as sunrpc's cache.
Just exports sunrpc's codes, and remove nfsd's redundant codes.
v8, same as v6
Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
1) The kernel sunrpc code needs to handle seconds since epoch
greater than 2147483647. This means functions that parse time
as an int need to handle it as time_t.
2) The kernel changes must be accompanied by userspace changes
in nfs-utils.
Signed-off-by: Harshula Jayasuriya <harshula@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
commit d202cce896
sunrpc: never return expired entries in sunrpc_cache_lookup
moved the 'entry is expired' test from cache_check to
sunrpc_cache_lookup, so that it happened early and some races could
safely be ignored.
However the ip_map (in svcauth_unix.c) has a separate single-item
cache which allows quick lookup without locking. An entry in this
case would not be subject to the expiry test and so could be used
well after it has expired.
This is not normally a big problem because the first time it is used
after it is expired an up-call will be scheduled to refresh the entry
(if it hasn't been scheduled already) and the old entry will then
be invalidated. So on the second attempt to use it after it has
expired, ip_map_cached_get will discard it.
However that is subtle and not ideal, so replace the "!cache_valid"
test with "cache_is_expired".
In doing this we drop the test on the "CACHE_VALID" bit. This is
unnecessary as the bit is never cleared, and an entry will only
be cached if the bit is set.
Reported-by: Bodo Stroesser <bstroesser@ts.fujitsu.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
It is possible for a race to set CACHE_PENDING after cache_clean()
has removed a cache entry from the cache.
If CACHE_PENDING is still set when the entry is finally 'put',
the cache_dequeue() will never happen and we can leak memory.
So set a new flag 'CACHE_CLEANED' when we remove something from
the cache, and don't queue any upcall if it is set.
If CACHE_PENDING is set before CACHE_CLEANED, the call that
cache_clean() makes to cache_fresh_unlocked() will free memory
as needed. If CACHE_PENDING is set after CACHE_CLEANED, the
test in sunrpc_cache_pipe_upcall will ensure that the memory
is not allocated.
Reported-by: <bstroesser@ts.fujitsu.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Passing this pointer is redundant since it's stored on cache_detail structure,
which is also passed to sunrpc_cache_pipe_upcall () function.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This callback will allow to simplify upcalls in further patches in this
series.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Commit bbf43dc888 "sunrpc/cache.h: replace
simple_strtoul" introduced new range-checking which could cause get_int
to fail on unsigned integers too large to be represented as an int.
We could parse them as unsigned instead--but it turns out svcgssd is
actually passing down "-1" in some cases. Which is perhaps stupid, but
there's nothing we can do about it now.
So just revert back to the previous "sloppy" behavior that accepts
either representation.
Cc: stable@vger.kernel.org
Reported-by: Sven Geggus <lists@fuchsschwanzdomain.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
I don't think there's a practical difference for the range of values
these interfaces should see, but it would be safer to be unambiguous.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This patch replaces the usage of simple_strtoul with kstrtoint in
get_int(), since the simple_str* family doesn't account for overflow
and is deprecated.
Also, in this specific case, the long from strtol is silently converted
to an int by the caller.
As Joe Perches <joe@perches.com> suggested, this patch also removes
the redundant temporary variable rv, since kstrtoint() will not write to
anint unless it's successful.
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Eldad Zack <eldad@fogrefinery.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Neaten code style in get_int().
Also use sizeof() instead of hard coded number as suggested by
Joe Perches <joe@perches.com>.
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Eldad Zack <eldad@fogrefinery.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
All cache users now uses network-namespace-aware routines, so generic ones
are obsolete.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Acked-by: J. Bruce Fields <bfields@redhat.com>
This patch prepares infrastructure for network namespace aware cache detail
allocation.
One note about adding network namespace link to cache structure. It's going to
be used later in NFS DNS cache parsing routine (nfs_dns_parse for rpc_pton()
call).
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Acked-by: J. Bruce Fields <bfields@redhat.com>
This precursor patch splits SUNRPC cache creation and PipeFS registartion.
It's required for latter split of NFS DNS resolver cache creation per network
namespace context and PipeFS registration/unregistration on MOUNT/UMOUNT
events.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As promised in feature-removal-schedule.txt it is time to
remove the nfsctl system call.
Userspace has perferred to not use this call throughout 2.6 and it has been
excluded in the default configuration since 2.6.36 (9 months ago).
So this patch removes all the code that was being compiled out.
There are still references to sys_nfsctl in various arch systemcall tables
and related code. These should be cleaned out too, probably in the next
merge window.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
* 'for-2.6.38' of git://linux-nfs.org/~bfields/linux: (62 commits)
nfsd4: fix callback restarting
nfsd: break lease on unlink, link, and rename
nfsd4: break lease on nfsd setattr
nfsd: don't support msnfs export option
nfsd4: initialize cb_per_client
nfsd4: allow restarting callbacks
nfsd4: simplify nfsd4_cb_prepare
nfsd4: give out delegations more quickly in 4.1 case
nfsd4: add helper function to run callbacks
nfsd4: make sure sequence flags are set after destroy_session
nfsd4: re-probe callback on connection loss
nfsd4: set sequence flag when backchannel is down
nfsd4: keep finer-grained callback status
rpc: allow xprt_class->setup to return a preexisting xprt
rpc: keep backchannel xprt as long as server connection
rpc: move sk_bc_xprt to svc_xprt
nfsd4: allow backchannel recovery
nfsd4: support BIND_CONN_TO_SESSION
nfsd4: modify session list under cl_lock
Documentation: fl_mylease no longer exists
...
Fix up conflicts in fs/nfsd/vfs.c with the vfs-scale work. The
vfs-scale work touched some msnfs cases, and this merge removes support
for that entirely, so the conflict was trivial to resolve.
Remove kobject.h from files which don't need it, notably,
sched.h and fs.h.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NeilBrown <neilb@suse.de>
[bfields@redhat.com: moved svcauth_unix_purge outside ifdef's.]
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Existing calls do the same, but for the init_net.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Being a hash table, hlist is the best option.
There is currently some ugliness were we treat "->next == NULL" as
a special case to avoid having to initialise the whole array.
This change nicely gets rid of that case.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
The current practice of waiting for cache updates by queueing the
whole request to be retried has (at least) two problems.
1/ With NFSv4, requests can be quite complex and re-trying a whole
request when a later part fails should only be a last-resort, not a
normal practice.
2/ Large requests, and in particular any 'write' request, will not be
queued by the current code and doing so would be undesirable.
In many cases only a very sort wait is needed before the cache gets
valid data.
So, providing the underlying transport permits it by setting
->thread_wait,
arrange to wait briefly for an upcall to be completed (as reflected in
the clearing of CACHE_PENDING).
If the short wait was not long enough and CACHE_PENDING is still set,
fall back on the old approach.
The 'thread_wait' value is set to 5 seconds when there are spare
threads, and 1 second when there are no spare threads.
These values are probably much higher than needed, but will ensure
some forward progress.
Note that as we only request an update for a non-valid item, and as
non-valid items are updated in place it is extremely unlikely that
cache_check will return -ETIMEDOUT. Normally cache_defer_req will
sleep for a short while and then find that the item is_valid.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This protects us from confusion when the wallclock time changes.
We convert to and from wallclock when setting or reading expiry
times.
Also use seconds since boot for last_clost time.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Rather can duplicating this idiom twice, put it in an inline function.
This reduces the usage of 'expiry_time' out side the sunrpc/cache.c
code and thus the impact of a change that is about to be made to that
field.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
This patch makes the cache_cleaner workqueue deferrable, to prevent
unnecessary system wake-ups, which is very important for embedded
battery-powered devices.
do_cache_clean() is called every 30 seconds at the moment, and often
makes the system wake up from its power-save sleep state. With this
change, when the workqueue uses a deferrable timer, the
do_cache_clean() invocation will be delayed and combined with the
closest "real" wake-up. This improves the power consumption situation.
Note, I tried to create a DECLARE_DELAYED_WORK_DEFERRABLE() helper
macro, similar to DECLARE_DELAYED_WORK(), but failed because of the
way the timer wheel core stores the deferrable flag (it is the
LSBit in the time->base pointer). My attempt to define a static
variable with this bit set ended up with the "initializer element is
not constant" error.
Thus, I have to use run-time initialization, so I created a new
cache_initialize() function which is called once when sunrpc is
being initialized.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
For events that are rare, such as referral DNS lookups, it makes limited
sense to have a daemon constantly listening for upcalls on a channel. An
alternative in those cases might simply be to run the app that fills the
cache using call_usermodehelper_exec() and friends.
The following patch allows the cache_detail to specify alternative upcall
mechanisms for these particular cases.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This field is set once and never used; probably some artifact of an
earlier implementation idea.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Newer server features such as nfsv4 and gss depend on proc to work, so a
failure to initialize the proc files they need should be treated as
fatal.
Thanks to Andrew Morton for style fix and compile fix in case where
CONFIG_NFSD_V4 is undefined.
Cc: Andrew Morton <akpm@linux-foundation.org>
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
There's really nothing much the caller can do if cache unregistration
fails. And indeed, all any caller does in this case is print an error
and continue. So just return void and move the printk's inside
cache_unregister.
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Speed up high call-rate workloads by caching the struct ip_map for the peer on
the connected struct svc_sock instead of looking it up in the ip_map cache
hashtable on every call. This helps workloads using AUTH_SYS authentication
over TCP.
Testing was on a 4 CPU 4 NIC Altix using 4 IRIX clients, each with 16
synthetic client threads simulating an rsync (i.e. recursive directory
listing) workload reading from an i386 RH9 install image (161480 regular files
in 10841 directories) on the server. That tree is small enough to fill in the
server's RAM so no disk traffic was involved. This setup gives a sustained
call rate in excess of 60000 calls/sec before being CPU-bound on the server.
Profiling showed strcmp(), called from ip_map_match(), was taking 4.8% of each
CPU, and ip_map_lookup() was taking 2.9%. This patch drops both contribution
into the profile noise.
Note that the above result overstates this value of this patch for most
workloads. The synthetic clients are all using separate IP addresses, so
there are 64 entries in the ip_map cache hash. Because the kernel measured
contained the bug fixed in commit
commit 1f1e030bf7
and was running on 64bit little-endian machine, probably all of those 64
entries were on a single chain, thus increasing the cost of ip_map_lookup().
With a modern kernel you would need more clients to see the same amount of
performance improvement. This patch has helped to scale knfsd to handle a
deployment with 2000 NFS clients.
Signed-off-by: Greg Banks <gnb@melbourne.sgi.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We can now make some code static.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Neil Brown <neilb@suse.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
.. it makes some of the code nicer.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Cache_fresh is now only used in cache.c, so unexport it.
Part of cache_fresh (setting CACHE_VALID) should really be done under the
lock, while part (calling cache_revisit_request etc) must be done outside the
lock. So we split it up appropriately.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This has been replaced by more traditional code.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The C++-like 'template' approach proves to be too ugly and hard to work with.
The old 'template' won't go away until all users are updated.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These were an unnecessary wart. Also only have one 'DefineSimpleCache..'
instead of two.
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When registering an RPC cache, cache_register() always sets the owner as the
sunrpc module. However, there are RPC caches owned by other modules. With
the incorrect owner setting, the real owning module can be removed potentially
with an open reference to the cache from userspace.
For example, if one were to stop the nfs server and unmount the nfsd
filesystem, the nfsd module could be removed eventhough rpc.idmapd had
references to the idtoname and nametoid caches (i.e.
/proc/net/rpc/nfs4.<cachename>/channel is still open). This resulted in a
system panic on one of our machines when attempting to restart the nfs
services after reloading the nfsd module.
The following patch adds a 'struct module *owner' field in struct
cache_detail. The owner is further assigned to the struct proc_dir_entry
in cache_register() so that the module cannot be unloaded while user-space
daemons have an open reference on the associated file under /proc.
Signed-off-by: Bruce Allan <bwa@us.ibm.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
NeilBrown